twitt

Data Bite #01: Are we ready to embrace virtual schools?

Our Data Bites series provides digestible, bite-sized insights on a current issue by taking a close look at some key statistics. In this Data Bite, we revisit survey results from our 2019 'Digital Lessons' report, to examine how parents felt about the idea of virtual schools before COVID-19.

Background

Until a few weeks ago, virtual schooling was not a reality for most Australian parents. Now, the shutdown of school campuses in the wake of COVID-19 has thrust millions of Australian students into emergency forms of online education.

This abrupt switch-over has come as a shock to many parents and has left educators unsure of how online school arrangements will fare over the coming weeks and months. While we have some indication of the difficulties encountered by full-time virtual school students in the US, there is little or no empirical evidence to guide us towards a full-scale switch-over to online schooling in Australia.

Key Findings

  • Prior to the COVID-19 school shutdowns, most of the general public, parents and non-parents alike, did not approve of the idea of virtual schooling. Care must be taken to ensure that the compromised nature of the current switch-over to virtual schooling arrangements does not exacerbate opposition to future uses of online learning.
     
  • Parents of current school-aged children were more likely to approve of virtual schooling than those without children at school. This difference is likely to be generational and can be seen as an encouraging finding for those seeking to promote schools’ future use of online learning.
     
  • Parents’ pre-shutdown views about virtual schooling were not notably patterned in terms of socio-economic status, household income, educational background, and/or speaking a language other than English. We might expect the idea of virtual schooling and online lessons to appeal to certain groups of parents, such as those with better home resources. However, our analysis found that this was not the case.
     
  • Parents’ approval of virtual schooling was positively correlated with their overall perceptions of how well schools were doing. On one hand this is to be expected – parents who think schools are doing a good job trust them to continue doing so using virtual platforms. However, it does raise the issue of how the current switch-over to virtual school arrangements is working for those parents who were already disaffected with their children’s schooling.

A Note on the Data Source

The Digital Lessons report published the results of a survey of public opinion on the use of digital technologies in Australian schools, from a nationally-representative sample of 2052 Australian adults who were eligible to vote. Of the participants in the survey, 469 reported that they had school-aged children living at home with them. We re-examined the responses of these parents to a survey item that explored their attitudes towards different types of emerging technology that we thought were likely to come to prominence in schools during the 2020s. 

For the purpose of this new interrogation of the data, we have labelled the respondents who had school-aged children living at home with them as ‘parents’ and grouped all other respondents together as ‘non-parents’. We then re-examined the responses of the parents group to a survey item that explored their attitudes towards different types of emerging technology that we thought were likely to come to prominence in schools during the 2020s. We also contrasted their overall attitudes with those of the 1583 non-parents.

Of particular interest is the variable ‘approval of virtual schools’, to which we applied various types of statistical analyses as our main dependent measure. Virtual schools were described as fully online schools where students are taught entirely through the internet.

Headline Patterns and Trends

What are the overall levels of support for virtual schooling?

The majority of both parents and non-parents disapproved of the idea of virtual schooling. However, the numerical data (Table 1) raised further questions about the nature of the relationship between virtual schools and parental status. While the majority of respondents in both groups expressed disapproval of virtual schooling, it is evident that there are proportionately more parents than non-parents among those who supported virtual schooling.

Table 1. Approval of virtual schools by parental status

Approval of virtual schools

Non-parents

Parents

n

%

n

%

Strongly Disapprove

439

27.70

120

25.60

Disapprove

450

28.40

117

24.90

Not sure

405

25.60

95

20.30

Approve

236

14.90

99

21.10

Strongly Approve

53

3.30

38

8.10

Total N

1583

469

In order to see how the two variables of ‘parental status’ and ‘approval of virtual schools’ were related, we employed a series of standard statistical procedures. Results from both the Chi-square difference test [χ²(4,N=2052)=33.374, p< 0.001] and the Kruskal-Wallis test [χ² with ties (1,N=2052)=10.523, p=0.0012] were significant and indicated that parents and non-parents had different attitudes towards virtual schools. Furthermore, Spearman rank correlation also suggested a statistically significant, albeit weak, relationship between these two measures (Spearman's rho =0.0716, p<0.001).

These analyses alone do not fully explain any relationship between parental status and approval for virtual schooling. This is because Chi-square tends to increase in large samples in social science research contexts. Nor does a single binary correlation (that reveals a significant yet rather weak association between variables) explain the exact nature of their relationship. Therefore, we extended our analysis to examine how parental views of virtual schooling were linked to other social background measures.

Are there differences in parents’ support for virtual schooling?

Rather than considering the relationship between virtual schools and parental status in a social vacuum, we turned to a range of social background variables also included in the survey instrument that might potentially influence attitudes towards virtual schooling. We extended our analyses to include eight additional variables: socioeconomic status (SES), age, education, income, satisfaction with the public-school system, geographic location, gender, and a language other than English being spoken at home (Table 2).

Table 2. List of all variables with their measurement properties

Variables

Categories

Meaning

Coding

Missing cases

virtual schools

strongly disagree

disagree

not sure

agree

strongly agree

approval of virtual schools, fully online schools where all students are taught entirely through the internet

1-5

0

SES

low

medium

high

socioeconomic status (postcode data)

1-3

0

age

18–29

30–39

40–49

50–59

60+

age group

1-5

0

education

year12 or below

TAFE or technical

certificate or diploma

university degree or higher

highest level of education completed

1-4

35

income

up to $39.99k

40k–79.99k

80k–119.99k

120k–199.99k

200k+

income groups

1-5

308

school satisfaction

strongly disagree

disagree

not sure

agree

strongly agree

current rating of Australian public-schools' performance in terms of learning outcomes

1-5

0

location

metro

rural

geographical location of residence

0-1

0

gender

male

female

gender

0-1

6

other language

English only

speak a language other than English at home

speak a language other than English at home

0-1

25

parental status

not having school aged children living at home

having school aged children living at home

family status based on whether or not there are school-aged children in the home

0-1

0

We wanted to find out whether the inclusion of these additional social background measures could help identify the types of parents who were more or less likely to express approval for virtual schooling. We explored the two Spearman correlation tables from both parental status subsamples to map the relationships between approval of virtual schooling and the eight variables outlined above (Table 3).

Table 3. Spearman's correlations (rho) among all variables by parental status as grouping variable

virtual schools

SES

age

education

income

school rating

location

gender

other language

virtual schools

correlation

1

-0.023

-0.241

0.062

0.088

0.179

-0.018

-0.132

0.148

sig

0.399

0

0.025

0.001

0

0.506

0

0

SES

correlation

-0.083

1

-0.022

0.196

0.24

0.04

-0.39

-0.023

0.076

sig

0.095

0.426

0

0

0.151

0

0.41

0.006

age

correlation

-0.116

0.014

1

-0.184

-0.271

-0.157

0.143

0.005

-0.153

sig

0.019

0.785

0

0

0

0

0.856

0

education

correlation

0.07

0.148

-0.073

1

0.272

0.03

-0.167

-0.028

0.162

sig

0.158

0.003

0.144

0

0.279

0

0.314

0

income

correlation

-0.064

0.198

-0.032

0.393

1

0.094

-0.152

-0.099

0.086

sig

0.198

0

0.527

0

0.001

0

0

0.002

school satisfaction

correlation

0.214

-0.018

-0.217

0.078

0.084

1

-0.034

-0.03

0.043

sig

0

0.712

0

0.119

0.09

0.213

0.281

0.118

location

correlation

0.07

-0.327

0.1

-0.323

-0.195

-0.075

1

0.026

-0.148

sig

0.163

0

0.044

0

0

0.131

0.353

0

gender

correlation

-0.113

0.026

-0.083

-0.139

-0.146

-0.031

0.072

1

-0.037

sig

0.023

0.608

0.097

0.005

0.003

0.535

0.151

0.181

other language

correlation

0.106

0.007

-0.112

0.245

-0.04

0.101

-0.154

-0.059

1

sig

0.033

0.893

0.024

0

0.424

0.043

0.002

0.234

Below the diagonal: Spearman's correlations from the parents subsample (N=404 after listwise deletion); above the diagonal: Spearman's correlations from the non-parents subsample (N=1318 after listwise deletion)

This analysis revealed two positive and two negative significant associations in the parents group between approval of virtual schooling and the eight explanatory variables. Similarly, there are four positive and two negative significant associations in the non-parent subsample.

  • Among parents, age (rho= -0.116, p=0.019) and gender (rho= -0.113, p=0.023) were negatively related to attitudes towards virtual schooling. Being satisfied with the public-school system (rho=0.214, p<0.001) and speaking a language other than English at home (rho= 0.106, p=0.033) were positively related to attitudes towards virtual schooling.
     
  • Among non-parents, we found positive significant associations between attitudes towards virtual schooling and being satisfied with the public-school system (rho=0.179, p<0.001), speaking a language other than English at home (rho=0.148, p<0.001), income (rho=0.088, p=0.001), and education (rho=0.062, p=0.025). For non-parents, there were significant negative relations between attitudes towards virtual schooling and age (rho= -0.241, p<0.001) and gender (rho= -0.132, p<0.001).

Does having children at school affect respondents’ views of virtual schooling before the pandemic?

While simple exploratory techniques such as Chi-square difference testing and correlations are informative to some extent, they are not suited for the complex analytical task at hand. In order to assess the relative explanatory strength of the social background variables and parental status on respondents’ approval of virtual schooling, we conducted three ordinal logit regression analyses in a sequence: total sample, non-parents and then parents (Table 4).

Table 4. Ordered logistic regression models in the total, non-parents and parents samples

Effects on approval of virtual schools

total sample (N=1722)

non-parents (N=1318)

parents (N=404)

Odds Ratio

[95% Conf. Interval]

sig. (p)

Odds Ratio

[95% Conf. Interval]

sig. (p)

Odds Ratio

[95% Conf. Interval]

sig. (p)

school satisfaction

1.51

[1.341–1.695]

p<0.001

1.48

[1.291–1.689]

p<0.001

1.62

[1.269–2.063]

p<0.001

gender

0.62

[0.524–0.741]

p<0.001

0.61

[0.504–0.749]

p<0.001

0.63

[0.436–0.897]

0.011

location

1.18

[0.953–1.470]

0.13

1.09

[0.850–1.388]

0.51

1.66

[1.035–2.655]

0.035

income

0.98

[0.903–1.068]

0.67

1.03

[0.933–1.128]

0.6

0.83

[0.691–0.996]

0.045

age

0.79

[0.744–0.844]

p<0.001

0.79

[0.737–0.843]

p<0.001

0.82

[0.660–1.007]

0.058

SES

0.91

[0.807–1.033]

0.15

0.91

[0.786–1.043]

0.17

0.94

[0.724–1.208]

0.61

education

1.03

[0.956–1.110]

0.44

1.003

[0.923–1.091]

0.93

1.18

[0.992–1.398]

0.062

other language

1.67

[1.319–2.105]

p<0.001

1.82

[1.376–2.412]

p<0.001

1.29

[0.843–1.986]

0.239

parental status

1.23

[0.994–1.517]

0.057

Missing cases have been deleted listwise. Significant effects are in bold. Independent variables are listed in order of decreased significance in the parents sample. Total sample model fit: LR chi2(9)=196.49, p<0.001, Pseudo R2=0.038. Non-parents sample model fit: LR chi2(8)=155.68, p<0.001, Pseudo R2=0.0402. Parents sample model fit: LR chi2(8)=40.19, p<0.001, Pseudo R2=0.0319. All analyses have been carried out in Stata 16.

First, we regressed approval of virtual schooling on the nine explanatory variables (eight social background measures plus parental status). When parental status was entered into the analyses alongside the other social background measures, its effect on approval of virtual schooling became insignificant (p=0.057). Such effect-demotion was likely due to the presence of other, relatively stronger and more robust variables in the analysis.

Four variables were found to have much stronger and significant effects on approval of virtual schooling than parental status.

  • Respondents who spoke a language other than English at home were more likely (OR=1.67, p<0.001) to approve of virtual schooling when compared with those who spoke only English.
     
  • Respondents who were satisfied with the public-school system were more likely (OR=1.51, p<0.001) to approve of virtual schooling than those who had less favourable views on the school system.
     
  • The odds ratio of gender is below one (OR=0.62, p<0.001), which can be interpreted as a negative effect. That is, women were less likely to approve of virtual schooling than men.
     
  • The odds ratio of age is also below one (OR=0.79, p<0.001), indicating that older respondents are less likely than younger adults to approve of virtual schooling.

So what differences, if any, might be found between the two parental groups with respect to approval of virtual schooling? We undertook the two additional ordinal logistic regression models in a largely identical way to the process described above, except we regressed approval of virtual schooling separately for the non-parents and the parents groups on the eight social background variables only.

The results highlighted a few notable points. In the parents subgroup, it was only ‘satisfaction with the public-school system’ that had a very robust and highly significant effect on attitudes towards virtual schooling, dominating all other variables. While three other variables remained significant in this subgroup, their corresponding p-values approached the significance threshold, which indicates their instability. For example, although barely significant, the negative effect of income in this subgroup suggested that financially well-off parents were less keen on the idea of virtual schooling. One possible explanation for this difference is that this group may expect schools to provide a more comprehensive education experience that is only possible in a physical school.

It is also interesting to note that the ‘other language’ variable was no longer significant (p=0.239) in the regression on the parents subsample, although the original bivariate Spearman correlation was significant (rho=0.106, p=0.033). Furthermore, in both the overall regression analysis and the regression conducted with the non-parents subgroup, we found a considerable ‘other language’ effect (OR=1.67, p<0.001 and OR=1.82, p<0.001, respectively) on attitudes towards virtual schooling. However, this appeared to be no longer the case in the parents subgroup.

It is not immediately clear how to explain such effect-erosion. It may be that virtual schooling appeared attractive to other language speakers only if they were non-parents. For those in the parents group who spoke another language, it might be that satisfaction with the school system was a salient consideration in their views on virtual schooling and outweighed the relative importance of other effects included in our analyses.

So What? Points for Further Discussion

At the time of the survey, none of our respondents could have imagined the imminent nature of children being switched-over to fully online schooling. However, their responses do indicate the types of parent who might be more positive (or perhaps less resistant) towards the prospect of virtual schooling.

Looking towards the future, these responses also give us some indication of how schools can work with parents to successfully make use of online learning in the aftermath of the current COVID-19 shutdown. Indeed, schools’ current emergency adoption of online teaching and learning might well hasten the more measured adoption of virtual schooling approaches in the future. These technologies look set to play an increasing role in school education over the next ten years and beyond.

Although limited by what was asked in the original survey, the following points for discussion emerged from our analysis of the data.

  • Most parents do not approve of the idea of virtual schooling. It is important to be aware that the inevitably compromised and frustrating nature of the current switch-over to virtual schooling arrangements may exacerbate parents’ opposition to future uses of online education. Once regular schooling has resumed, education authorities and other education technology groups need to use the collective experience of the COVID-19 shutdown as a starting-point for working with parents, teachers and students to imagine better ways of using these technologies.
     
  • That said, we also found that parents of current school-aged children are more likely to approve of virtual schooling than those without children at school. This can be seen as an encouraging finding for those seeking to promote schools’ future use of online education. This difference might be generational – parents in our survey were predominantly aged between 30 and 49 years old, whereas the largest group of respondents without school-aged children were 60 or older. As such, it might be that approval for online classes increases over time, as we see new generations of parents who have experienced online education (at school or university) for themselves.
     
  • We might expect the idea of virtual schooling and online lessons to appeal to certain groups of parents, for example, those with better home resourcing. However, our analysis found that parents’ views about virtual schooling are not notably patterned in terms of SES, household income, educational background, or households where languages other than English are spoken. While this is encouraging, it will be important to make sure that the experiences of the current coerced switch-over to virtual schooling does not alter this.
     
  • Although not prominent and consistent in our analyses, there were indications of possible differences in approval for virtual schooling based on parents’ gender and their location (metropolitan vs regional areas). These issues need to be explored further.
     
  • Finally, we found that parents’ approval of virtual schooling and online lessons is positively correlated with their overall perceptions of how well schools were doing. On one hand this is to be expected – parents who think schools are doing a good job trust them to continue doing so using virtual platforms. However, it does point to the need for schools to be especially mindful of how the current switch-over to virtual lessons is working for those parents who were already disaffected with their children’s schooling. In the longer term, these are parent groups that need to be prioritized in future discussions of how schools can make use of these technologies.
No comment added yet.
VIEW MORE COMMENTS View fewer comments (0)

Republish

Republish

You may republish this article online or in print under our Creative Commons licence. You may not edit or shorten the text, you must attribute the article to Monash Education Futures, and you must include the author’s name in your republication.

If you have any questions, please email education.futures@monash.edu

https://monasheducationfuture.contentgrowthstack.io/republishing-guidelines

Title
Data Bite #01: Are we ready to embrace virtual schools?

Content

Content copied successfully
×

Subscribe

  • !
  • !
  • !
Expect approximately one email a month.
Privacy Policy

Thanks for subscribing

Check your inbox

We’ve sent an email to you